No full-text available
To read the full-text of this research,
you can request a copy directly from the authors.
That's More Like It: Multiple Exemplars Facilitate Word Learning
Abstract
Previous research indicates learning words facilitates categorisation. The current study explores how categorisation affects word learning. In the current study, we investigated whether learning about a category facilitates retention of newly learned words by presenting 2-year-old children with multiple referent selection trials to the same object category. In Experiment 1, children either encountered the same exemplar repeatedly or encountered multiple exemplars across trials. All children did very well on the initial task; however, only children who encountered multiple exemplars retained these mappings after a short delay. Experiment 2 replicated and extended this finding by exploring the effect of within-category variability on children's word retention. Children encountered either narrow or broad exemplars across trials. Again, all children did very well on the initial task; however, only children who encountered narrow exemplars retained mappings after a short delay. Overall, these data offer strong evidence that providing children with the opportunity to compare across exemplars during fast mapping facilitates retention.
... A large body of work demonstrates the benefits of comparing multiple objects (i.e., considering two or more instances simultaneously) versus labeling a single object for category learning and generalization (Gentner & Namy, 1999;Graham, Namy, Gentner, & Meagher, 2010;Medin, Goldstone, & Gentner, 1993;Namy & Clepper, 2010;Namy & Gentner, 2002;Namy, Smith, & Gershkoff-Stowe, 1997;Spalding & Ross, 2000;Twomey, Ranson, & Horst, 2014;Vukatana, Graham, Curtin, & Zepeda, 2015). For example, providing learners with opportunities to compare multiple instances has been shown to result in higher levels of exhaustive classification (Namy et al., 1997), increased ability to disregard irrelevant perceptual similarity in favor of relational matches (Gentner, Anggoro, & Klibanoff, 2011), accelerated verb learning (Childers, Paik, Flores, & Lai, 2017), facilitated solving of complex relational problems (Dixon & Bangert, 2004), and increased likelihood or reasoning through analogy (Gick & Holyoak, 1983). ...
... Multiple examples may also be more beneficial when learners must retain word-object pairings over a delay (Twomey et al., 2014). Recent research suggests that despite the fact that children can readily link a word to its referent in fast mapping studies, they appear to have difficulty in retaining that link over short time delays. ...
... However, when children were asked to select the referent after a 5-min delay, they responded at chance levels, suggesting that children failed to retain the object-label link over the short delay. Similar work that included multiple exemplars of a category found that children were able to generalize over a delay when given multiple examples but not when given a single example (Twomey et al., 2014). In addition, an eye-tracking study (Bion, Borovsky, & Fernald, 2013) found that children had difficulty in retaining a label after a delay when training involved a single example. ...
A large body of research indicates that children can map words to categories and generalize the label to new instances of the category after hearing a single instance of the category labeled. Additional research demonstrates that word learning is enhanced when children are presented with multiple instances of a category through comparison or contrast. In this study, 3-year-old children participated in a novel noun generalization task in which a label was given for either (a) a single instance of a category, (b) multiple instances of a category, or (c) contrasting a category instance with non-category members. Children were asked to extend the label to a new category at test either immediately (Study 1) or after a 10-s delay (Study 2). The results indicate that when tested immediately, children who heard a single instance labeled outperformed children who were presented with multiple instances. However, when tested after a brief delay, there was no difference among the conditions.
... In a series of 3D object word learning tasks, Twomey, Ranson, and Horst (2014) trained 30-monthold toddlers with novel word-object mappings and demonstrated that children retained words only after encountering multiple perceptually similar exemplars of each novel object category. We reasoned that 10-month-old infants may likewise benefit from exposure to multiple category exemplars. ...
... Thus, infants' failure to learn word-category associations in Experiment 2 is unlikely to stem from an inability to categorize. Rather, the lack of retention of these associations may be due to the increased cognitive load of encoding a category plus a label in Experiment 2 versus an individual object and a label in Experiment 1. Notably, this pattern of findingsretention after the same exemplar repeatedly but failure to retain after multiple exemplarsis the opposite of Twomey et al. (2014) findings for toddlers. We return to this issue in the General Discussion. ...
... When provided with richer perceptual input via multiple exemplars during training in Experiment 2, infants failed to learn labels. This finding diverges from the effect of variability on toddlers' word learning (Twomey et al., 2014), and from tasks which have shown the benefit of multiple exemplars in other domains (e.g., Kovack-Lesh & Oakes. 2007;Quinn & Bhatt, 2010;Rost & McMurray, 2009). ...
In this series of experiments, we tested the limits of young infants’ word learning and generalization abilities in light of recent findings reporting sophisticated word learning abilities in the first year of life. Ten-month-old infants were trained with two word-object pairs and tested with either the same or different members of the corresponding categories. In Experiment 1, infants showed successful learning of the word-object associations, when trained and tested with a single exemplar from each category. In Experiment 2, infants were presented with multiple within-category items during training but failed to learn the word-object associations. In Experiment 3, infants were presented with a single exemplar from each category during training, and failed to generalize words to a new category exemplar. However, when infants were trained with items from perceptually and conceptually distinct categories in Experiment 4, they showed weak evidence for generalization of words to novel members of the corresponding categories. It is suggested that word learning in the first year begins as the formation of simple associations between words and objects that become enriched as experience with objects, words and categories accumulates across development.
... Twomey et al. [29] suggest, however, that there are limits to the effects of variability on learning: although 30-month-olds learned labels for categories when the objects varied in colour (but not when they were in identical colours), children did not learn category labels when objects varied in shape and colour. Thus, too much variability disrupted children's learning of category labels. ...
... They found that children only showed target recognition in the variable condition which led them to assume that decontextualization helped the children to form strong word representations. Nonetheless, they also suggest that increased variability might disrupt successful learning and that this might relate to the learning environment: reduced variability might help in rich learning environments and increased background variability may boost learning in simpler learning environments (see [29,32]). Thus, the effect of variability may vary across contexts and guide attention allocation in different ways, thereby influencing learning behaviour (see [33]). ...
... We interpret these findings to suggest that redundant information from different domains supports the forming of rich lexical representations, but only if this information highlights the wordobject association and does not distract from it, and only towards the third year of life. This is in line with the literature suggesting a beneficial effect of consistency on word learning [21], in contrast to a beneficial effect of variability on generalization over different members of a category [29,47,48]. Nevertheless, some recent work suggests an impact of lower-level variability (variability in the colour of the background on which objects were presented) on learning of word-object associations [31]. ...
Communication with young children is often multimodal in nature, involving, for example, language and actions. The simultaneous presentation of information from both domains may boost language learning by highlighting the connection between an object and a word, owing to temporal overlap in the presentation of multimodal input. However, the overlap is not merely temporal but can also covary in the extent to which particular actions co-occur with particular words and objects, e.g. carers typically produce a hopping action when talking about rabbits and a snapping action for crocodiles. The frequency with which actions and words co-occurs in the presence of the referents of these words may also impact young children’s word learning. We, therefore, examined the extent to which consistency in the co-occurrence of particular actions and words impacted children’s learning of novel word–object associations. Children (18 months, 30 months and 36–48 months) and adults were presented with two novel objects and heard their novel labels while different actions were performed on these objects, such that the particular actions and word– object pairings always co-occurred (Consistent group) or varied across trials (Inconsistent group). At test, participants saw both objects and heard one of the labels to examine whether participants recognized the target object upon hearing its label. Growth curve models revealed that 18-month-olds did not learn words for objects in either condition, and 30-month-old and 36- to 48-month-old children learned words for objects only in the Consistent condition, in contrast to adults who learned words for objects independent of the actions presented. Thus, consistency in the multimodal input influenced word learning in early childhood but not in adulthood. In terms of a dynamic systems account of word learning, our study shows how multimodal learning settings interact with the child’s perceptual abilities to shape the learning experience.
... Other evidence appears to run counter to this finding. Twomey, Ranson, and Horst (2014) examined the effects of encountering variable exemplars on 2.5-year-old children's gradual, long-term word learning. In their initial experiment, children learned nonword labels when the same object exemplar was presented repeatedly or when multiple object exemplars that differed in color were presented. ...
... However, the 2.5-year-old children only learned in the one context condition and did not benefit from context variability, suggesting a developmental effect for input variability. These results are similar to those of Twomey et al. (2014), who found that too much variability did not aid learning at this young age. Because the objects used in training for the Vlach and Sandhofer (2011) study already varied in color, texture, and perceptual features, perhaps varying the context as well was too much variability for these young children. ...
... The current findings complement prior work by others for typically developing children (Perry et al., 2010;Twomey et al., 2014) and extend them to a clinical population characterized by language deficits. However, it is important to note the differences between studies, particularly that of Perry et al. (2010), which was the most similar to the current study. ...
Purpose:
Variability in the input plays an important role in language learning. The current study examined the role of object variability for new word learning by preschoolers with specific language impairment (SLI).
Method:
Eighteen 4- and 5-year-old children with SLI were taught 8 new words in 3 short activities over the course of 3 sessions. Half of the children saw 3 identical objects corresponding to each new word during training (No Variability group); the other half of the children saw 3 different objects corresponding to each new word during training (High Variability group). Children completed vocabulary learning tests for objects seen during training and for new within-category objects that were never seen during training as a test of category generalization. Learning was assessed the day after each training activity, and retention was assessed 3 weeks after the last training session.
Results:
There were no group differences on trained or generalization items immediately following training sessions. However, children in the High Variability group demonstrated significantly better retention 3 weeks after experimental training.
Conclusion:
These findings demonstrate that object variability facilitates retention of new word learning by children with SLI.
Supplemental material:
https://doi.org/10.23641/asha.5583979.
... For example, visual variability encountered across target stimuli facilitates categorization in 6-to 7-month-old infants (Quinn & Bhatt, 2010), and phonological variability in affect or speaker has been shown to support early word recognition (Rost & McMurray, 2009;Singh, 2008). Recent work has shown that target variability also affects word learning: in a referent selection task, when shown a novel 3D object category with exemplars that varied in color, 30-month-old children learned category labels, but did not when exemplars were identical, or varied in shape and color simultaneously (Twomey, Ranson, & Horst, 2014; see also Ankowski, Vlach & Sandhofer, 2013;Perry, Samuelson, Malloy & Schiffer, 2010). Thus, while some target variability supports language learning, too much target variability appears Running head: BACKGROUND VARIABILITY HELPS WORD LEARNING 5 to disrupt it. ...
... Where's the apple). In line with typical 3D object referent selection tasks in which warm-up trials include ostensive Running head: BACKGROUND VARIABILITY HELPS WORD LEARNING 10 feedback (e.g., Twomey et al., 2014), during the next 3 s the target object rotated accompanied by a twinkling sound, followed by ostensive auditory feedback (e.g., There's the apple!). In the final 1 s the objects bounced diagonally towards the bottom right hand corner and offscreen, accompanied by the sound of children cheering. ...
... The variable color condition is one of the first word learning experiments to show children performing apparently poorly during referent selection but successfully at test. In typical analyses of word learning tasks with 3D objects, test trials for which children have not correctly mapped novel labels during referent selection are excluded, with the rationale that children do not learn (correct) novel label-novel object mappings when (incorrectly) mapping novel labels to known objects (e.g., Hilton & Westermann, 2016;Twomey et al., 2014). From this perspective it is surprising that children in the variable color condition were the only ones to show robust retention: if during training these children were not looking to novel objects at above-chance levels, how then did they retain? ...
Variability is prevalent in early language acquisition, however whether it supports or hinders learning is unclear: while target variability has been shown to facilitate word learning, variability in competitor items has been shown to make the task harder. Here we tested whether background variability could boost learning in a referent selection task. Two groups of two-year-old children saw arrays of one novel and two known objects on a screen, and heard a novel or known label. Stimuli were identical across conditions, with the exception that in the constant color condition objects appeared on a uniform white background, and in the variable color condition backgrounds were different, uniform colors. At test, only children in the variable condition showed evidence of retaining label-object associations. These data support findings from the adult memory literature, which suggest that variability supports learning by decontextualizing representations. We argue that these data are consistent with dynamic systems accounts of learning in which low-level entropy adds sufficient noise to the developmental system to precipitate a change in behavior.
... It is therefore ideal that visual stimuli not have been seen before, in order to ensure that any inferences made regarding learning were not actually due to participants' exposure to the items prior to the experiment. For many experimental designs it is important that objects also be easy to distinguish from each other (e.g., Twomey, Ranson, & Horst, 2014;Yu & Smith, 2007); however, for other designs Electronic supplementary material The online version of this article (doi:10.3758/s13428-015-0647-3) contains supplementary material, which is available to authorized users. it can be useful to have objects that are somewhat similar (e.g., Homa et al., 2011;Hout & Goldinger, 2015). ...
... This database was originally created for use in word learning experiments, primarily with children. However, researchers may also require novel objects when investigating categorization (e.g., Twomey et al., 2014), (visual) short-term memory (e.g., Kwon, Luck, & Oakes, 2014), and long-term visual memory (e.g., Hout & Goldinger, 2010. This newly-developed set of photographs is freely available to the scientific community from the authors for noncommercial use. ...
... Stimuli Participants saw three exemplars for ten of the object categories seen in Experiments 1 and 2. We selected exemplars that only differed from each other in color, since this is a common method for forming categories for experimental research (Golinkoff, Hirsh-Pasek, Bailey, & Wenger, 1992;Twomey et al., 2014;Vlach, Sandhofer, & Kornell, 2008;Woodward, Markman, & Fitzsimmons, 1994). These images were created in a fashion identical to that in Experiment 1. ...
Many experimental research designs require images of novel objects. Here we introduce the Novel Object and Unusual Name (NOUN) Database. This database contains 64 primary novel object images and additional novel exemplars for ten basic- and nine global-level object categories. The objects’ novelty was confirmed by both self-report and a lack of consensus on questions that required participants to name and identify the objects. We also found that object novelty correlated with qualifying naming responses pertaining to the objects’ colors. Results from a similarity sorting task (and subsequent multidimensional scaling analysis on the similarity ratings) demonstrated that the objects are complex and distinct entities that vary along several featural dimensions beyond simply shape and color. A final experiment confirmed that additional item exemplars comprise both sub- and superordinate categories. These images may be useful in a variety of settings, particularly for developmental psychology and other research in language, categorization, perception, visual memory and related domains.
... For example, in a longitudinal training study, Perry and colleagues demonstrated that training 18-month-old children with variable exemplars of categories of objects supported them in more successfully generalizing new words to unfamiliar exemplars, and led to increases in vocabulary development relative to peers who saw the same object repeatedly (Perry et al., 2010). Further research has shown that two-year-old children who were taught novel category names by exposing them to variable exemplars of objects from the same category were able to retain more novel category names than children who were exposed to single exemplars only (Twomey et al., 2014). Thus, exposing children to multiple, variable exemplars rather than only one category member may provide children with the opportunity to compare similar features of each of the exemplars, and, therefore, facilitate word learning for that category. ...
... Thus, exposing children to multiple, variable exemplars rather than only one category member may provide children with the opportunity to compare similar features of each of the exemplars, and, therefore, facilitate word learning for that category. However, when exemplar variability was too high, that is, exemplars differed on multiple characteristics (shape, size, colour, etc), word learning was hindered (Twomey et al., 2014). Twomey and colleagues explained this finding by suggesting that when exemplar variability is high, more attentional resources are required, leaving fewer for committing the word-object associations to memory. ...
Although strong claims have been made about museums being ideal word learning environments, these are yet to be empirically supported. In the current study, 152 four‐ to five‐year‐olds children (81‐M, 71‐F) from minority backgrounds were taught six vocabulary items either in a museum, in their classroom with museum resources, or in their classroom with classroom resources. At test, children taught in their classroom with museum resources produced significantly more correct responses than children taught in the museum or in their classroom with classroom resources. Children were also significantly better at retaining the target vocabulary items than recalling them. These data demonstrate how context can impact word learning and point to the benefits of a collaborative relationship between schools and museums to support children's language development.
... Early knowledge transfer literature suggested that increasing the number and variability of exemplars can benefit learning and generalization (Gick & Holyoak, 1987 for a summary). Similarly, the developmental literature (e.g., Aguilar et al., 2018;Perry et al., 2010;Twomey et al., 2014) and applied behavioral analysis (e.g., Stokes & Baer, 1977;Stokes & Osnes, 2016;Swan et al., 2016) recommend the use of multiple exemplars to promote learning and generalization. Based on some of these findings, Thompson (1989) suggested that training multiple examples might promote generalization in aphasia. ...
... In contrast, in the high exemplar variability condition, each item was trained using three picture exemplars. This was intended to increase stimuli variability and promote within-level generalization (e. g., Aguilar et al., 2018;Perry et al., 2010;Twomey et al., 2014). In the verbal description condition, each item was trained by presenting a short verbal description prompt in both written and auditory formats. ...
Introduction
: There is a pressing need to improve computer-based treatments for aphasia to increase access to long-term effective evidence-based interventions. The current single case design incorporated two learning principles, adaptive distributed practice and stimuli variability, to promote acquisition, retention, and generalization of words in a self-managed computer-based anomia treatment.
Methods
: Two participants with post-stroke aphasia completed a 12-week adaptive distributed practice naming intervention in a single-case experimental design. Stimuli variability was manipulated in three experimental conditions: high exemplar variability, low exemplar variability, and verbal description prompt balanced across 120 trained words. Outcomes were assessed at 1-week, 1-month, and 3-months post-treatment. Statistical comparisons and effect sizes measured in the number of words acquired, generalized, and retained were estimated using Bayesian generalized mixed-effect models.
Results
: Participants showed large and robust acquisition, generalization, and retention effects. Out of 120 trained words, participant 1 acquired ∼77 words (trained picture exemplars) and ∼63 generalization words (untrained picture exemplars of treated words). Similarly, participant 2 acquired ∼57 trained words and ∼48 generalization words. There was no reliable change in untrained control words for either participant. Stimuli variability did not show practically meaningful effects.
Conclusions
: These case studies suggest that adaptive distributed practice is an effective method for re-training more words than typically targeted in anomia treatment research (∼47 words on average per Snell et al., 2010). Generalization across experimental conditions provided evidence for improved lexical access beyond what could be attributed to simple stimulus-response mapping. These effects were obtained using free, open-source flashcard software in a clinically feasible, asynchronous format, thereby minimizing clinical implementation barriers. Larger-scale clinical trials are required to replicate and extend these effects.
... Variability among instances of visual forms is a known driver of category learning (e.g. Perry, Samuelson, Malloy, & Schiffer, 2010;Twomey, Lush, Pearce, & Horst, 2014;Twomey, Ranson, & Horst, 2014). Compared to typeface letters, handwritten letters are variable in form-each production of a letter is different from the last-especially when produced by young children (Longstaff & Heath, 1997;Wing & Nimmo-Smith, 1987). ...
... This visual input may be responsible for the changes in ventral-temporal function after letter production. Ventral-temporal cortex is broadly associated with object categorization processes (for review, see Grill-Spector & Weiner, 2014), and the development of object categorization processes is largely driven by the perceptual differentiation that follows exposure to category variability (Li & James, 2016;Perry et al., 2010;Twomey, Lush, et al., 2014;Twomey, Ranson, et al., 2014). Our results suggest that ventral-temporal cortex may be most sensitive to the variability present in handwritten forms when children are first learning about letters and that this sensitivity to visual variability may be a part of how ventral-temporal cortex undergoes developmental changes that contribute to the formation of category-specific responses. ...
Letter production through handwriting creates visual experiences that may be important for the development of visual letter perception. We sought to better understand the neural responses to different visual percepts created during handwriting at different levels of experience. Three groups of participants, younger children, older children, and adults, ranging in age from 4.5 to 22 years old, were presented with dynamic and static presentations of their own handwritten letters, static presentations of an age‐matched control’s handwritten letters, and typeface letters during fMRI. First, data from each group were analyzed through a series of contrasts designed to highlight neural systems that were most sensitive to each visual experience in each age group. We found that younger children recruited ventral‐temporal cortex during perception and this response was associated with the variability present in handwritten forms. Older children and adults also recruited ventral‐temporal cortex; this response, however, was significant for typed letter forms but not variability. The adult response to typed letters was more distributed than in the children, including ventral‐temporal, parietal, and frontal motor cortices. The adult response was also significant for one’s own handwritten letters in left parietal cortex. Second, we compared responses among age groups. Compared to older children, younger children demonstrated a greater fusiform response associated with handwritten form variability. When compared to adults, younger children demonstrated a greater response to this variability in left parietal cortex. Our results suggest that the visual perception of the variability present in handwritten forms that occurs during handwriting may contribute to developmental changes in the neural systems that support letter perception.
... Variability among instances of visual forms is a known driver of category learning (e.g. Perry, Samuelson, Malloy, & Schiffer, 2010;Twomey, Lush, Pearce, & Horst, 2014;Twomey, Ranson, & Horst, 2014). Compared to typeface letters, handwritten letters are variable in form-each production of a letter is different from the last-especially when produced by young children (Longstaff & Heath, 1997;Wing & Nimmo-Smith, 1987). ...
... This visual input may be responsible for the changes in ventral-temporal function after letter production. Ventral-temporal cortex is broadly associated with object categorization processes (for review, see Grill-Spector & Weiner, 2014), and the development of object categorization processes is largely driven by the perceptual differentiation that follows exposure to category variability (Li & James, 2016;Perry et al., 2010;Twomey, Lush, et al., 2014;Twomey, Ranson, et al., 2014). Our results suggest that ventral-temporal cortex may be most sensitive to the variability present in handwritten forms when children are first learning about letters and that this sensitivity to visual variability may be a part of how ventral-temporal cortex undergoes developmental changes that contribute to the formation of category-specific responses. ...
... In contrast, 10-month-old infants in a categorization task formed a robust category when familiarized with novel stimuli in an order that maximized, but not minimized, overall perceptual differences between successive stimuli (Mather & Plunkett, 2011). Still other studies have uncovered a "Goldilocks" effect in which learning is optimal when stimuli are of intermediate predictability (Kidd et al., 2012(Kidd et al., , 2014; see also Kinney & INFANT CURIOSITY: A NEUROCOMPUTATIONAL APPROACH 9 Kagan, 1976;Twomey, Ranson, & Horst, 2014). From this perspective, the degree of novelty and/or complexity in the environment that best supports learning is unclear. ...
... best support learning (Kidd et al., 2012(Kidd et al., , 2014Kinney & Kagan, 1976;Twomey et al., 2014). Equally, simplicity has been shown to support learning in some cases (Bulf et al., 2011;Son, Smith, & Goldstone, 2008). ...
Infants are curious learners who drive their own cognitive development by imposing structure on their learning environment as they explore. Understanding the mechanisms by which infants structure their own learning is therefore critical to our understanding of development. Here we propose an explicit mechanism for intrinsically motivated information selection that maximizes learning. We first present a neurocomputational model of infant visual category learning, capturing existing empirical data on the role of environmental complexity on learning. Next we “set the model free”, allowing it to select its own stimuli based on a formalization of curiosity and three alternative selection mechanisms. We demonstrate that maximal learning emerges when the model is able to maximize stimulus novelty relative to its internal states, depending on the interaction across learning between the structure of the environment and the plasticity in the learner itself. We discuss the implications of this new curiosity mechanism for both existing computational models of reinforcement learning and for our understanding of this fundamental mechanism in early development.
... First, we introduced a condition in which infants were familiarized to multiple exemplars of a category. Research across a number of domains has demonstrated that variability promotes a comparative process that allows young children to attend to critical information and thereby facilitates the learning and generalization of information (Casasola, 2005;Childers, 2011;Childers & Paik, 2009;Twomey, Ranson, & Horst, 2014). Similarly, comparison allows young children to detect commonalities among objects that may not be otherwise detected when categorizing (Gentner & Namy, 1999Namy & Gentner, 2002;Namy, Gentner, & Clepper, 2007). ...
... For example, in the domain of language, variation from different sources (e.g., speaker identify, affective quality) augments infants' phonological processing (e.g., Rost & McMurray, 2009) and facilitates their ability to detect words from a continuous speech stream (Houston & Jusczyk, 2000;Singh, 2008). Furthermore, the opportunity to compare across multiple variable exemplars can assist young children in acquiring the meaning of novel verbs (Childers & Paik, 2009) and spatial terms (Casasola, 2005), in retaining newly learnt name-object mappings (Twomey et al., 2014), and in generalizing novel labels to novel exemplars (Perry, Samuelson, Malloy, & Schiffer, 2010). Similarly, within the categorization domain, presentation with multiple exemplars has been shown to facilitate the formation of categories for both familiar and novel objects (Gentner & Namy, 1999Namy & Gentner, 2002;Namy et al., 2007;Oakes & Ribar, 2005). ...
Across three experiments, we examined 9- and 11-month-olds' mappings of novel sound properties to novel animal categories. Infants were familiarized with novel animal–novel sound pairings (e.g., Animal A [red]–Sound 1) and then tested on: (1) their acquisition of the original pairing and (2) their generalization of the sound property to a new member of a familiarized category (e.g., Animal A [blue]–Sound 1). When familiarized with a single exemplar of a category, 11-month-olds showed no evidence of acquiring or generalizing the animal–sound pairings. In contrast, 11-month-olds learnt the original animal–sound mappings and generalized the sound property to a novel member of that category when familiarized with multiple exemplars of a category. Finally, when familiarized with multiple exemplars, 9-month-old infants learnt the original animal–sound pairing, but did not extend the novel sound property. The results of these experiments provide evidence for developmental differences in the facilitative role of multiple exemplars in promoting the learning and generalization of information.
... Recent research demonstrates that category variability influences toddlers' noun learning (Perry, Samuelson, Malloy, & Schiffer, 2010;Twomey, Ranson & Horst, 2013) while labelling and vocabulary level affect object categorisation (Plunkett, Hu & Cohen, 2008;Gershkoff-Stowe & Smith, 2004). However, how such findings relate to verb learning remains unclear. ...
... Children in the variable actions condition dishabituated to the out-of-category action paired with the habituated verb. This finding is consistent with the literature on noun learning, where visual variability helps young children form object categories, label them with nouns, and generalise those nouns to new exemplars (e.g., Perry et al. 2010;Twomey, Ranson & Horst, 2013). Similarly, the developmental categorisation literature suggests that such visual variability triggers object comparison and draws attention to category-relevant features whilst decreasing attention to categoryirrelevant features (e.g., Oakes, Plumert, Lansink & Merryman, 1996;Kovack-Lesh & Oakes, 2007;Quinn & Bhatt, 2010). ...
Research demonstrates that within-category visual variability facilitates noun learning; however, the effect of visual variability on verb learning is unknown. We habituated 24-month-old children to a novel verb paired with an animated star-shaped actor. Across multiple trials children either saw a single action from an action category (identical actions condition, e.g., travelling while repeatedly changing into a circle shape) or multiple actions from that action category (variable actions condition, e.g., travelling while changing into a circle shape, then a square shape, then a triangle shape) or a. Four test trials followed habituation. One paired the habituated verb with a new action from the habituated category (e.g., “dacking” + pentagon shape) and one with a completely novel action (e.g., “dacking” + leg movement). The others paired a new verb with a new same-category action (e.g., “keefing” + pentagon shape), or a completely novel category action (e.g., “keefing” + leg movement). Although all children discriminated novel verb/action pairs, children in the identical actions condition discriminated trials that included the completely novel verb, while children in the variable actions condition discriminated the out-of-category action. These data suggest that—as in noun learning—visual variability affects verb learning and children’s ability to form action categories.
... Early categorization skills are strongly implicated in the ability to acquire the labels of referents (Gelman & Markman, 1986;Markman, 1989;Twomey et al., 2014). Categorization skills have not been extensively studied across the autism spectrum and across dimensions of language ability. ...
Word learning requires successful pairing of form and meaning. A common hypothesis about the process of word learning is that initially, infants work on identifying the phonological segments corresponding to words (speech analysis), and subsequently map those segments onto meaning. A range of theories have been proposed to account for the underlying mechanisms and factors in this remarkable achievement. While some are mainly concerned with the sensorimotor affordances and perceptual properties of referents out in the world, other theories emphasize the importance of language as a system, and the relations among language units (other words or syntax). Recent approaches inspired by neuro-science suggest that the storage and processing of word meanings is supported by neural systems subserving both the representation of conceptual knowledge and its access and use (Lambon Ralph et al., Nature Reviews Neuroscience 18:42–55, 2017). Developmental disorders have been attested to impact on different aspects of word learning. While impaired word knowledge is not a hallmark of Autism Spectrum Disorder (ASD), and remains largely understudied in this population, there is evidence that there are, sometimes subtle, problems in that domain, reflected in both how such knowledge is acquired and how words are used (Vulchanova et al., Word knowledge and word usage: A cross-disciplinary guide to the mental lexicon, Mouton De Gruyter, 2020). In addition, experimental evidence suggests that children with autism present with specific problems in categorizing the referents of linguistic labels leading to subsequent problems with using those labels (Hartley and Allen, Autism 19:570–579, 2015). Furthermore, deficits have been reported in some of the underlying mechanisms, biases and use of cues in word learning, such as e.g., object shape (Field et al., Journal of Autism and Developmental Disorders 46:1210–1219, 2016; Tek et al., Autism Research 1:208–222, 2008). Finally, it is likely that symbol use might be impaired in ASD, however, the direction of the causal relationship between social and communication impairment in autism and symbolic skills is still an open question (Allen and Lewis, Journal of Autism and Developmental Disorders 45:1–3, 2015; Allen and Butler, British Journal of Developmental Psychology 38:345–362, 2020; Wainwright et al., Journal of Autism and Developmental Disorders 50:2941–2956, 2020). Further support for impaired symbol formation in autism comes from the well-attested problems with figurative, non-literal language use (e.g., metaphors, idioms, hyperbole, irony) (Vulchanova et al., Frontiers in Human Neuroscience 9:24, 2015). Here we propose that embodied theories of cognition which link perceptual experience with conceptual knowledge (see Eigsti, Frontiers in Psychology 4:224, 2013; Klin et al., Philosophical Transactions of the Royal Society of London. Series B: Biological Sciences 358:345–360, 2003) might be useful in explaining the difficulty in symbolic understanding that individuals with autism face during the word learning process.
... There are similar findings in the second language literature, where there is evidence that encountering varying talkers can assist in auditory and speech perception tasks and robust vocabulary learning (Strange and Dittmann, 1984;Logan et al., 1991;Lively et al., 1993;Clopper and Pisoni, 2004;Barcroft and Sommers, 2005;Sinkeviciute et al., 2019). More opportunity to compare across multiple exemplars has also been found to boost memory of novel words in child language: In Twomey et al. (2014), 2-year-olds presented with multiple exemplars of a category demonstrated better retention of label-object associations than children who only encountered the same exemplar repeatedly. The authors argue that children who encountered multiple exemplars retained words at greater rates because the act of comparing exemplars across trials enabled children to identify category-relevant features while downplaying categoryirrelevant features. ...
Human language is characterized by productivity, that is, the ability to use words and structures in novel contexts. How do learners acquire these productive systems? Under a discriminative learning approach, language learning involves using cues to predict and discriminate linguistic outcomes and “generalization” involves dissociating idiosyncratic irrelevant cues in favour of informative, invariant cues. The current work tests the predictions of this account using the learning of spatial adpositions as a test case. Spatial adpositions describe the location of one object in relation to another (e.g., English prepositions “above'' and “below'') and may occur in reversible sentences, such as "the picture is above the window"; generalization involves using these terms in novel contexts, such as with unattested nouns. Computational simulations implementing an error-driven, discriminative learning process, demonstrate that broadening the irrelevant cues associated with the stimuli may boost the discovery of invariant cues, i.e., the association between the adposition and the spatial relation. We explore the predictions of these models in human learners by adapting a training paradigm introduced by Hsu and Bishop (2014) to teach typically-developing 7/8 year olds spatial adpositions in an unfamiliar language (Japanese) using a computerized learning game. We manipulate the cue variability by comparing groups of children trained with more variable sentences (high variability) with a condition with repetition of the same sentences (low variability). A third condition (skew) tests whether learning and generalization are boosted when learning from a heavy tailed distribution that more closely resembles that of natural language. We will examine the following predictions: (1) for sentences with novel nouns, participants trained with variable sentences will show better performance (i.e., stronger generalization) than those trained with repeated sentences; (2) in contrast, those trained with repeated sentences will show stronger performance in training itself (i.e., stronger item learning); (3) training with a heavy tailed distribution -- more closely resembling the natural one -- will lead to the strongest item learning and generalization.
... This view is also supported by research on noun learning and categorization. An exposure to variable exemplars helps children learn and retain novel labels (Perry, Samuelson, Malloy, & Schiffer, 2010;Twomey, Ranson, & Horst, 2014). In addition, seeing a broad range of exemplars facilitates children's formation and generalization of nonlinguistic and linguistic categories of geometric shapes (Bomba & Siqueland, 1983), categories of objects (Gentner & Namy, 1999;Graham, Namy, Gentner, & Meagher, 2010;Oakes & Spalding, 1997), perceptual organization (Bhatt & Quinn, 2011), and non-adjacent dependencies of 3-element strings (e.g., aXc or bXd - Gomez, 2002). ...
Verbs serve as the architectural centerpiece of sentences, making verb learning pivotal for language acquisition. Verb learning requires both the formation of a verb-action mapping and the abstraction of relations between an object and its action. Two competing positions have been proposed to explain the process of verb learning: (a) seeing a highly variable range of exemplars allows children to detect and abstract the commonalities across actions—the action invariants; and (b) seeing a less variable range of exemplars enables children to focus on and extract the action invariants. Using manner—a major component of verb meaning in English—as a test case, this study addressed this debate by examining the influence of manner variability on the ability to fast-map new verbs and extend them to novel exemplars in 2-, 3-, and 4-year-old English-speaking children. Results contribute to this debate by showing that high manner variability hindered fast-mapping but facilitated extension to manner variations in the 2- and 3-year-olds. Thus, high exemplar variability may affect verb fast-mapping and extension differently. Furthermore, manner variability did not affect 4-year-olds’ (or adults’) fast-mapping or extension, suggesting that the influence of exemplar variability on verb learning attenuates with age. Finally, manner variability did not affect agent or object extension, revealing a component-specific effect of exemplar variability on verb extension.
... In general, high variability can make learning more difficult when learners are in the very early stages of acquiring a target behavior. For instance, beginners and children who are only just 'getting the hang' of a motor skill (e.g., a tennis serve) or who are just being familiarized with a novel category benefit from low variability during initial practice (e.g., blocked training as opposed to more variable interleaved training; exposure to exemplars with little to no variation between them), and may experience difficulties or even get overwhelmed when too much variability is introduced at first [55,71,[74][75][76][77][78][79]. For example, students with less prior knowledge who are learning to solve math problems benefit from receiving less variable examples first, while the opposite is true for students with more prior knowledge [63,80]. ...
Learning is using past experiences to inform new behaviors and actions. Because all experiences are unique, learning always requires some generalization. An effective way of improving generalization is to expose learners to more variable (and thus often more representative) input. More variability tends to make initial learning more challenging, but eventually leads to more general and robust performance. This core principle has been repeatedly rediscovered and renamed in different domains (e.g., contextual diversity, desirable difficulties, variability of practice). Reviewing this basic result as it has been formulated in different domains allows us to identify key patterns, distinguish between different kinds of variability, discuss the roles of varying task-relevant versus irrelevant dimensions, and examine the effects of introducing variability at different points in training.
... However, in real life, people choose to obtain information from more than one person. Furthermore, many studies have shown that children with multiple learning samples have better word learning abilities compared to those who learn from only one sample [36][37][38]. For this reason, we choose GB and RB i as the two examples of the ith particle social learning part in this study. ...
In the traditional particle swarm optimization algorithm, the particles always choose to learn from the well-behaved particles in the population during the population iteration. Nevertheless, according to the principles of particle swarm optimization, we know that the motion of each particle has an impact on other individuals, and even poorly behaved particles can provide valuable information. Based on this consideration, we propose Lévy flight-based inverse adaptive comprehensive learning particle swarm optimization, called LFIACL-PSO. In the LFIACL-PSO algorithm, First, when the particle is trapped in the local optimum and cannot jump out, inverse learning is used, and the learning step size is obtained through the Lévy flight. Second, to increase the diversity of the algorithm and prevent it from prematurely converging, a comprehensive learning strategy and Ring-type topology are used as part of the learning paradigm. In addition, use the adaptive update to update the acceleration coefficients for each learning paradigm. Finally, the comprehensive performance of LFIACL-PSO is measured using 16 benchmark functions and a real engineering application problem and compared with seven other classical particle swarm optimization algorithms. Experimental comparison results show that the comprehensive performance of the LFIACL-PSO outperforms comparative PSO variants.
... The importance of perceptual variability during learning is not a new hypothesis. Indeed, according to several studies, comparison would play a critical role in the categorization of novel objects (e.g., Gentner and Namy, 1999;Namy and Gentner, 2002;Graham et al., 2010;Twomey et al., 2014). Some studies even suggested that the greater the variability among exemplars during learning, the better the generalization to new category instances (Posner and Keele, 1968;Perry et al., 2010). ...
Based on evidence that learning new characters through handwriting leads to better recognition than learning through typing, some authors proposed that the graphic motor plans acquired through handwriting contribute to recognition. More recently two alternative explanations have been put forward. First, the advantage of handwriting could be due to the perceptual variability that it provides during learning. Second, a recent study suggests that detailed visual analysis might be the source of the advantage of handwriting over typing. Indeed, in that study, handwriting and composition –a method requiring a detailed visual analysis but no specific graphomotor activity– led to equivalent recognition accuracy, both higher than typing. The aim of the present study was to assess whether the contribution of detailed visual analysis is observed in preschool children and to test the variability hypothesis. To that purpose, three groups of preschool children learned new symbols either by handwriting, typing, or composition. After learning, children performed first a four-alternative recognition task and then a categorization task. The same pattern of results as the one observed in adults emerged in the four-alternative recognition task, confirming the importance of the detailed visual analysis in letter-like shape learning. In addition, results failed to reveal any difference across learning methods in the categorization task. The latter results provide no evidence for the variability hypothesis which would predict better categorization after handwriting than after typing or composition.
... consistencies to bootstrap audiovisual learning over multiple exposures. Our findings are in line with the regularity principle of statistical learning (Perry et al., 2010;Vlach and Sandhofer, 2011;Twomey et al., 2014), in which the cognitive system structures inherent environmental variability by integrating frequently occurring items by their co-occurrence, or consistency. This enables us to build supraordinate categories for words, and parts of words in the lexicon, and associated semantic webs. ...
Learning to read involves efficient binding of visual to auditory information. Aberrant cross-modal binding skill has been observed in both children and adults with developmental dyslexia. Here, we examine the contribution of episodic memory to acquisition of novel cross-modal bindings in typical and dyslexic adult readers. Participants gradually learned arbitrary associations between unfamiliar Mandarin Chinese characters and English-like pseudowords over multiple exposures, simulating the early stages of letter-to-letter sound mapping. The novel cross-modal bindings were presented in consistent or varied locations (i.e., screen positions), and within consistent or varied contexts (i.e., co-occurring distractor items). Our goal was to examine the contribution, if any, of these episodic memory cues (i.e., the contextual and spatial properties of the stimuli) to binding acquisition, and investigate the extent to which readers with and without dyslexia would differ in their reliance on episodic memory during the learning process. Participants were tested on their ability to recognize and recall the bindings both during training and then in post-training tasks. We tracked participants’ eye movements remotely with their personal webcams to assess whether they would re-fixate relevant empty screen locations upon hearing an auditory cue—indicative of episodic memory retrieval—and the extent to which the so-called “looking-at-nothing behavior” would modulate recognition of the novel bindings. Readers with dyslexia both recognized and recalled significantly fewer bindings than typical readers, providing further evidence of their persistent difficulties with cross-modal binding. Looking-at-nothing behavior was generally associated with higher recognition error rates for both groups, a pattern that was particularly more evident in later blocks for bindings encoded in the inconsistent location condition. Our findings also show that whilst readers with and without dyslexia are capable of using stimulus consistencies in the input—both location and context—to assist in audiovisual learning, readers with dyslexia appear particularly reliant on consistent contextual information. Taken together, our results suggest that whilst readers with dyslexia fail to efficiently learn audiovisual binding as a function of stimulus frequency, they are able to use stimulus consistency—aided by episodic recall—to assist in the learning process.
... While these are not necessarily explicit steps the child thinks about, they do represent a real-time competition in which the child must sort through the relevant and irrelevant information relatively quickly. This competition is driven by a number of factors including the number of competitors (Axelsson et al., 2012;Zosh, Brinster, & Halberda, 2013), novelty of the targets and foils (Horst, Samuelson, Kucker, & McMurray, 2011;Kucker, McMurray, & Samuelson, 2018), saliency of known foils (Pomper & Saffran, 2019), the child's current vocabulary knowledge (Bion, Borovsky, & Fernald, 2013;Kucker, McMurray, & Samuelson, 2020), pragmatic cues such as pointing or joint attention (Tomasello, Strosberg, & Akhtar, 1996), and the repetition (Horst, 2013) and variability (Twomey, Ranson, & Horst, 2014) of the items present. These factors can all be present in a variety of contexts, but what specific processes are prominent may vary across children and environments. ...
Word learning unfolds over multiple, cascading pathways which support in-the-moment processing and learning. The process is refined with each exposure to a word, and exposures to new words occur across a variety of forms and contexts. However, as children are exposed to more and more digital media, the ways in which children encounter, learn, and build on their vocabulary is also shifting. These shifts represent changes in context, content, and at the level of the child that can lead to negative outcomes. Less work, however, has discussed what these differences mean for how things change in the underlying developmental cascade and learning processes. Here, we suggest that the increasing presence of digital media may shift the developmental pathways for learning (the chain of events that support future learning) but not necessarily the developmental processes (the mechanisms underlying learning). Moreover, the interaction of the two may lead to different behavior and outcomes for learning in a digital era. We argue it is imperative for researchers to not only study how digital media differs from everyday learning, but directly measure if the well-worn pathways, processes, and variables found with decades of research with real items translate to a digital media era.
... "car") and potentially prompting shape-based generalization to additional exemplars (e.g. green cars; Allen, Hartley, & Cain, 2015Twomey, Ranson, & Horst, 2014). Exploring the efficacy and benefits of such training would be an interesting topic for future research. ...
... Des travaux récents offrent ainsi de mieux comprendre le rôle de l'intérêt pour la nouveauté et de l'intérêt pour l'inattendu comme moteurs d'apprentissage. Il a par exempleété observé que l'apprentissage d'une nouvelle catégorie d'objets chez des bébésâgés de 10 mois (Mather & Plunkett, 2011) ou d'un nouveau mot chez des bébésâgés de 2 ans (Twomey et al., 2014)était augmenté par une maximisation de la variabilité entre les items présentés aux bébés. Cette amélioration des capacités d'apprentissage pourrait, entre autres,être sous-tendue par une attention soutenue des bébés lors de la présentation d'items très différents les uns des autres, reflet de leur intérêt pour la nouveauté (Mather, 2013;Twomey & Westermann, 2018 (Begus et al., 2014), que des bébésâgés de 17 moisétaient plus attentifsà des situations leur permettant un progrès en apprentissage (ici l'apprentissage d'une nouvelle grammaire) qu'à des situations n'offrant aucun apprentissage (Gerken et al., 2011), ou encore que des bébésâgés de 20 mois demandaient de l'aideà un partenaire social lorsque cela leur permettait d'accéderà une information qu'ils ne possédaient pas pour résoudre un problème (ici, retrouver l'emplacement d'un objet caché) (Goupil et al., 2016, voir aussi Begus & Southgate, 2012, sur le rôle du pointage dans l'acquisition d'informations chez le bébé). ...
Au regard de la place croissante occupée par l'intelligence artificielle dans nos sociétés, la nécessité d'affiner notre compréhension de la notion d'apprentissage semble plus importante que jamais. S'intégrant dans une telle optique, cette thèse de doctorat cherche à mettre en lumière les mécanismes d'apprentissage permettant au bébé d'acquérir au cours de la première année de vie une utilisation appropriée et différenciée de son corps lui permettant d'interagir de manière efficace avec son environnement physique et social, ce que nous désignons ici par le terme de "savoir-fairecorporel". Le postulat au cœur de ce travail de recherche est le suivant : l'acquisition progressive du savoir-faire corporel au cours de la vie fœtale et des premiers mois de vie post-partum est sous-tendue par deux mécanismes d'apprentissage, l'exploitation de la sensibilité aux contingencessensorimotrices et la motivation intrinsèque. Ce travail de thèse explore la première partie de ce postulat, c'est-à-dire le rôle jouée par la sensibilité aux contingences sensorimotrices dans le développement du savoir-faire corporel. Afin d'investiguer cette hypothèse, ce travail de thèsese concentre dans un premier temps sur l'analyse critique des données expérimentales déjà existantes sur le sujet, à la fois en psychologie du développement et en robotique développementale. Dans un second temps, cette recherche vise à approfondir notre compréhension de l'acquisition du savoir-faire corporel à travers l'expérimentation chez le bébé âgé de moins d'un an.
... However, it is a more common phenomenon in human society that people always have multiple exemplars during their social learning process. Moreover, many studies indicate that children with multiple learning exemplars can offer a better word learning ability compared to those children who are only learning from one exemplar [2,4,37] . The simplest explanation is that an individual can extract much more useful knowledge from multiple information providers, i.e., multiple exemplars, than from a single exemplar. ...
... Despite being sensitive to visual context, younger children may also be less efficient than adults in extracting the relevant information, which may give rise to quantitative differences. Thus, younger children might require more contrast between shape pairs to extract relevant features (see Smith commentary in Marks et al., 1987) or might require more exemplars to extract or retain the relevant information (Casasola & Park, 2013;Twomey, Ranson, & Horst, 2014;cf. Maguire, Hirsh-Pasek, Golinkoff, & Brandone, 2008). ...
... In contrast, in reaching versions of the tasks 18-month-old toddlers do not perform well on reference selection or retention (Kucker, McMurray, & Samuelson, 2018), 24-month-olds perform well on reference selection, but they do not retain the mappings (Samuelson & Horst, 2008), and by 30 months of age children are good at both (see Kucker, McMurray, & Samuelson, 2015a for a review and discussion). Both reference selection and retention, however, can be influenced by a number of external and organismic factors (see e.g., Axelsson & Horst, 2014;Kalashnikova, Escudero, & Kidd, 2018;Kucker & Samuelson, 2012;Pomper & Saffran, 2018, Twomey, Ranson, & Horst, 2014. Still other research indicates that children as young as 13 months of age can map and, in some cases, retain novel word-object mappings when only one name and one object are presented at a time (Schafer & Plunkett, 1998;Woodward, Markman, & Fitzsimmons, 1994). ...
The use of global, standardized instruments is conventional among clini-cians and researchers interested in assessing neurocognitive development. Exclusively relying on these tests for evaluating effects may underestimate or miss specific effects on early cognition. The goal of this review is to identify alternative measures for possible inclusion in future clinical trials and interventions evaluating early neurocognitive development. The domains included for consideration are attention, memory, executive function , language, and socioemotional development. Although domain-based tests are limited, as psychometric properties have not yet been well-established, this review includes tasks and paradigms that have been reliably used across various developmental psychology laboratories. ARTICLE HISTORY
... For typically developing children acquiring language, it is not necessary to explain these similarities and differences. Mere exposure to multiple examples of category members and the category label is sufficient to induce the formation of a word class (Perry et al., 2010;Twomey, Ranson, & Horst, 2014;Vlach & Sandhofer, 2011). ...
Purpose:
Statistical learning research seeks to identify the means by which learners, with little perceived effort, acquire the complexities of language. In the past 50 years, numerous studies have uncovered powerful learning mechanisms that allow for learning within minutes of exposure to novel language input.
Method:
We consider the value of information from statistical learning studies that show potential for making treatment of language disorders faster and more effective.
Results:
Available studies include experimental research that demonstrates the conditions under which rapid learning is possible, research showing that these findings apply to individuals with disorders, and translational work that has applied learning principles in treatment and educational contexts. In addition, recent research on memory formation has implications for treatment of language deficits.
Conclusion:
The statistical learning literature offers principles for learning that can improve clinical outcomes for children with language impairment. There is potential for further applications of this basic research that is yet unexplored.
... Indeed, evidence suggests that word learning is particularly challenging for children when increasing amounts of perceptual information are presented. For example, children struggle to learn object names when target object categories are highly variable (Twomey, Ranson, & Horst, 2014), when target objects are presented in less predictable locations (Benitez & Smith, 2012), and with multiple combinations of extraneous objects, rather than the same combinations repeated (Axelsson & Horst, 2014). Such findings are consistent with cognitive load theory (Sweller, 1988(Sweller, , 1989, or see Paas, Renkl, & Sweller, 2003, for a review), which explains how working memory capacity is inherently limited and is especially problematic in situations with extraneous information. ...
Two experiments tested how the number of illustrations in storybooks influences 3.5-year-old children's word learning from shared reading. In Experiment 1, children encountered stories with two regular-sized A4 illustrations, one regular-sized A4 illustration, or one large-sized A3 illustration (in the control group) per spread. Children learned significantly fewer words when they had to find the referent within two illustrations presented at the same time. In Experiment 2, a gesture was added to guide children's attention to the correct page in the 2-illustration condition. Children who saw two illustrations with a guiding gesture learned words as well as children who had seen only one illustration per spread. Results are discussed in terms of the cognitive load of word learning from storybooks.
... For example, visual variability encountered across stimuli facilitates categorization in 6-to 7-month-old infants (Quinn & Bhatt, 2010), and phonological variability in affect or speaker has been shown to support early word recognition (Rost & McMurray, 2009). Recent work has revealed a similar effect of variability on word learning: when shown a novel 3D object category with exemplars that varied in color, 30-month-old children learned category labels, but did not when exemplars were identical, or varied in shape and color simultaneously (Twomey, Ranson, & Horst, 2014). Thus, while some target variability supports word learning, too much variability appears to disrupt it. ...
Variability is important in language acquisition; however, whether it supports or hinders learning is unclear: while 3D object studies suggest that children learn word-object mappings better when the object varies, storybook studies indicate that variability in the context in which new objects are shown impairs learning. We tested a dynamic systems account in which background variability should boost learning by speeding the emergence of new behaviors. Two groups of two-year-old children saw arrays of one novel and two known objects on a screen, and heard a novel or known label. Stimuli were identical across conditions, with the exception that in the constant condition objects appeared on a white background, and in the variable condition backgrounds were colored. At test, only children in the variable condition showed evidence of word learning. These data suggest that extraneous variability supports learning by decontextualizing representations, and indicate that low-level entropy adds sufficient noise to the developmental system to triggger a change in behavior.
... There were 12 referent categories: each included 1 referent exemplar (the prototype) and 1 referent exemplar that varied from the other one in color, size, or both (the non-prototype). Two referent exemplars for each category were used during training as some similarity and some variation within categories promotes both learning and generalization (Rost, 2011;Twomey et al., 2013). There was one generalization exemplar for each referent category used during testing that varied from the other two exemplars in color, size, or both. ...
Research on word learning has focused on children?s ability to identify a target object when given the word form after a minimal number of exposures to novel word-object pairings. However, relatively little research has focused on children?s ability to retrieve the word form when given the target object. The exceptions involve asking children to recall and produce forms, and children typically perform near floor on these measures. In the current study, 3- to 5-year-old children were administered a novel test of word form that allowed for recognition memory and manual responses. Specifically, when asked to label a previously trained object, children were given three forms to choose from: the target, a minimally different form, and a maximally different form. Children demonstrated memory for word forms at three post-training delays: 10 mins (short-term), 2?3 days (long-term), and 6 months to 1 year (very long-term). However, children performed worse at the very long-term delay than the other time points, and the length of the very long-term delay was negatively related to performance. When in error, children were no more likely to select the minimally different form than the maximally different form at all time points. Overall, these results suggest that children remember word forms that are linked to objects over extended post-training intervals, but that their memory for the forms gradually decreases over time without further exposures. Furthermore, memory traces for word forms do not become less phonologically specific over time; rather children either identify the correct form, or they perform at chance.
... Background: Previous research on referent selec7on has mainly inves6gated how proper6es of objects themselves affect children's ability to learn word--object mappings, such as the variability of novel targets (Twomey, Ranson, & Horst, 2014). However, while related research shows that background context affects infants' categoriza6on and noun generaliza6on (Goldenberg & Sandhofer, 2013;Vlach & Sandhofer, 2011), the effect of background context on referent selec6on is unknown. ...
Introduction In word learning studies children are presented with a novel object alongside two or more familiar competitor objects and are asked to point to one of the objects in response to a label (e.g. Which one’s the blicket?). Research shows that even young infants can reliably select a novel object in response to a novel label (Houston-Price, Plunkett & Harris, 2005). Previous research on referent selection has mainly investigated how properties of objects themselves, such as the number of known competitors or variability of novel targets, affect children’s ability to learn word-object mappings (e.g., Horst, Scott & Pollard, 2010; Twomey, Ranson & Horst, 2014). However, while related research shows that background context affects infants’ categorization and noun generalization (Vlach & Sandhofer, 2011; Goldenberg & Johnson, 2015; Goldenberg & Sandhofer, 2013), the effect of background context on referent selection is unknown. The current study therefore examined infants’ referent selection performance when objects were presented on single or variable-colored backgrounds. Method Two-year-old infants (N = 31) were presented with three combinations of novel label-object associations presented on a computer screen. Half of the infants were shown the objects on a white background (constant condition) and half on five different colored backgrounds (variable condition). Each trial consisted of one novel and two familiar objects side-by-side. Infants saw each combination five times (once with each background color in the variable condition) and were asked for each novel object three times (using a recording played through speakers) and every familiar stimulus once. An eye tracker measured infants’ looking toward the objects and the background. Results In trials where the novel target was named, infants in the constant condition looked at the novel target longer (M = 0.56, SD = 0.22) than infants in the variable condition (M = 0.48, SD = 0.25), t (283) = 2.85, p = .005, d = 0.17. In contrast, infants in the variable condition spent more time looking at the background (M = 0.29, SD = 0.6) than those in the constant condition (M = 0.19, SD = 0.46), t (455) = -1.90, p = .05, d = 0.09. Time course analysis of the proportion of target looking between the two groups revealed that looking preferences of the infants in the variable condition reverted back to chance levels (0.33) faster than in the constant condition. Discussion These findings highlight the differences between the processes involved in referent selection and generalization tasks. In category generalization tasks, infants encounter a single stimulus on each training trial, whereas the in referent selection tasks infants encounter multiple objects simultaneously. Referent selection tasks therefore require infants to process more information. While variability in generalization tasks help infants decontextualize, the changing backgrounds in the current study interfered with infants’ attention during the more complex referent selection task.
... The integration of novel words into long-term memory is clearly important for language acquisition and sleep appears to affect children's retention of novel words. However, another skill that is important for language development is the ability to generalize across instances (i.e., across different members of a category), so that children can flexibly extend their learning to other items and in other contexts (Waxman and Booth, 2000;Twomey et al., 2014). There is evidence to suggest that sleep can also affect participants' capacity to generalize their knowledge of multiple items (Darsaud et al., 2011). ...
In the first few years of life children spend a good proportion of time sleeping as well as acquiring the meanings of hundreds of words. There is now ample evidence of the effects of sleep on memory in adults and the number of studies demonstrating the effects of napping and nocturnal sleep in children is also mounting. In particular, sleep appears to benefit children's memory for recently-encountered novel words. The effect of sleep on children's generalization of novel words across multiple items, however, is less clear. Given that sleep is polyphasic in the early years, made up of multiple episodes, and children's word learning is gradual and strengthened slowly over time, it is highly plausible that sleep is a strong candidate in supporting children's memory for novel words. Importantly, it appears that when children sleep shortly after exposure to novel word-object pairs retention is better than if sleep is delayed, suggesting that napping plays a vital role in long-term word retention for young children. Word learning is a complex, challenging, and important part of development, thus the role that sleep plays in children's retention of novel words is worthy of attention. As such, ensuring children get sufficient good quality sleep and regular opportunities to nap may be critical for language acquisition.
... Importantly, there is substantial evidence that simply disambiguating the referent of a novel word once is not on its own sufficient for word learning (Bion, Borovsky, & Fernald, 2013;Horst & Samuelson, 2008;Kucker, McMurray, & Samuelson, 2015;Mather & Plunkett, 2009;McMurray, Horst, & Samuelson, 2012;Munro, Baker, McGregor, Docking, & Arculi, 2012;Twomey, Ranson & Horst, 2014). Rather, word learning is the result of incremental cross-situational learning, in which label-referent mappings are gradually strengthened via repeated encounters with the mapping (e.g., Fazly, Alishahi, & Stevenson, 2010;Smith & Yu, 2008;Yu & Smith, 2007;Yurovsky, Fricker, Yu, & Smith, 2014). ...
It is well-established that toddlers can correctly select a novel referent from an ambiguous array in response to a novel label. There is also a growing consensus that robust word learning requires repeated label-object encounters. However, the effect of the context in which a novel object is encountered is less wellunderstood. We present two embodied neural network replications of recent empirical tasks, which demonstrated that the context in which a target object is encountered is fundamental to referent selection and word learning. Our model offers an explicit account of the bottom-up associative and embodied mechanisms which could support children's early word learning and emphasises the importance of viewing behaviour as the interaction of learning at multiple timescales.
... Still other studies have uncovered a Goldilocks effect, in which learning is optimal when stimuli are of intermediate difficulty. In particular, a series of recent experiments showed that 8-month-old infants looked longer in response to auditory or visual stimuli that were moderately, but not maximally or minimally, predictable [25], [26], and older children in a hybrid word learning/categorization task retained novel label/object mappings when category exemplars were of intermediate perceptual variability [27]. Overall, then, it is not clear what degree of difficulty is optimal for infant categorization. ...
Infants are curious learners who drive their own cognitive development by imposing structure on their learning environments as they explore. Understanding the mechanisms underlying this curiosity is therefore critical to our understanding of development. However, very few studies have examined the role of curiosity in infants' learning, and in particular, their categorization; what structure infants impose on their own environment and how this affects learning is therefore unclear. The results of studies in which the learning environment is structured a priori are contradictory: while some suggest that complexity optimizes learning, others suggest that minimal complexity is optimal, and still others report a Goldilocks effect by which intermediate difficulty is best. We used an auto-encoder network to capture empirical data in which 10-month-old infants' categorization was supported by maximal complexity [1]. When we allowed the same model to choose stimulus sequences based on a " curiosity " metric which took into account the model's internal states as well as stimulus features, categorization was better than selection based solely on stimulus characteristics. The sequences of stimuli chosen by the model in the curiosity condition showed a Goldilocks effect with intermediate complexity. This study provides the first computational investigation of curiosity-based categorization, and points to the importance characterizing development as emerging from the relationship between the learner and its environment.
... Different tasks support different types of category learning. For example, yes/no questions lead to a stronger shape bias than forced-choice questions , various types of feedback differentially affect learning categories with highly salient features vs. less salient features (Hammer et al., 2012) and highly variable category members facilitate category name generalization (Perry et al., 2010) whereas less variable category members facilitate category name retention (Twomey et al., 2014). ...
Children develop in a real, messy world in which learning unfolds through time and space shared with others. Understanding how children develop in this complex environment will require a solid, theoretically-grounded understanding of how the child and environment interact-both within and beyond the laboratory. We as researchers understand the scientific value in testing children in carefully-controlled environments, but for our findings to have any impact on children's lives we must strive to understand how the processes we study in the lab operate in the real, busy environments in which children interact with peers and adults.
... Still other studies have uncovered a Goldilocks effect, in which learning is optimal when stimuli are of intermediate difficulty. In particular, a series of recent experiments showed that 8-month-old infants looked longer in response to auditory or visual stimuli that were moderately, but not maximally or minimally, predictable [25], [26], and older children in a hybrid word learning/categorization task retained novel label/object mappings when category exemplars were of intermediate perceptual variability [27]. Overall, then, it is not clear what degree of difficulty is optimal for infant categorization. ...
... Although fast mapping appears to be relatively easy, learning to remember the name-object associations is more difficult. Indeed, without pre-exposure to the objects (Kucker & Samuelson, 2012), explicit naming (Axelsson, Churchley, & Horst, 2012), or multiple trials per referent category (Smith & Yu, 2008;Twomey, Ranson, & Horst, 2014), young children fail to recall name-object associations after as little as 5 min (see also Bion, Borovsky, & Fernald, 2013;Gurteen, Horne, & Erjavac, 2011). Children also struggle to learn new words via fast mapping if they encounter referents among arrays with more than three familiar objects (Horst, Scott, & Pollard, 2010; see also Zosh, Brinster, & Halberda, 2013), or among arrays that include other novel objects for which they are also trying to learn new names (Axelsson & Horst, 2013;Wilkinson, Ross, & Diamond, 2003). ...
... Importantly, there is substantial evidence that simply disambiguating the referent of a novel word once is not on its own sufficient for word learning (Bion, WORD LEARNING IN A ROBOTIC SYSTEM 5 Borovsky, & Fernald, 2013;Horst & Samuelson, 2008;Kucker, McMurray, & Samuelson, 2015;Mather & Plunkett, 2009;McMurray, Horst, & Samuelson, 2012;Munro, Baker, McGregor, Docking, & Arculi, 2012;Twomey, Ranson & Horst, 2014). Rather, word learning is the result of incremental cross-situational learning, in which label-referent mappings are gradually strengthened via repeated encounters with the mapping (e.g., Fazly, Alishahi, & Stevenson, 2010;Smith & Yu, 2008;Yu & Smith, 2007;Yurovsky, Fricker, Yu, & Smith, 2014). ...
Children learn words with remarkable speed and flexibility. However, the cognitive basis of young children’s word learning is disputed. Further, although research demonstrates that children’s categories and category labels are interdependent, how children learn category labels is also a matter of debate. Recently, biologically plausible, computational simulations of children’s behavior in experimental tasks have investigated the cognitive processes that underlie learning. The ecological validity of such models has been successfully tested by deploying them in robotic systems (Morse, Belpaeme, Cangelosi, & Smith, 2010). The authors present a simulation of children’s behavior in a word learning task (Twomey & Horst, 2011) via an embodied system (iCub; Metta, et al., 2010), which points to associative learning and dynamic systems accounts of children’s categorization. Finally, the authors discuss the benefits of integrating computational and robotic approaches with developmental science for a deeper understanding of cognition.
Children in everyday environments experience verbs separated by minutes or hours and linked to events that vary in their similarity. Prior studies have shown that seeing similar events can be beneficial for verb learning (e.g., complex events), but there is also evidence that varied events or seeing both similar and varied events is useful; more studies are needed. In addition, few prior verb studies have tested verb learning from spaced practice. In Study 1, 3½- and 4½-year-olds (N = 72) saw either three similar events, three varied events, or a single live event (control) while hearing a new verb; events were separated by 1-min delays. Results showed better performance in multiple-event conditions than in the single-event condition and showed more extensions with age. Specifically, children benefitted more from seeing varied events with age. In Study 2, 2½-, 3½-, and 4½-year-olds (N = 163) either saw similar and then varied events or saw all varied video events separated by 1-min delays or no delays. The youngest children performed significantly better in the similar first condition than in the all varied condition, showing the first evidence of this benefit following spaced practice. In addition, as in Study 1, performance after seeing varied events increased with age. Together, these studies show that children can compare events separated in time and that their ability to learn verbs from varied examples develops with age.
In order to learn the mappings from words to referents, children must integrate co‐occurrence information across individually ambiguous pairs of scenes and utterances, a challenge known as cross‐situational word learning. In machine learning, recent multimodal neural networks have been shown to learn meaningful visual‐linguistic mappings from cross‐situational data, as needed to solve problems such as image captioning and visual question answering. These networks are potentially appealing as cognitive models because they can learn from raw visual and linguistic stimuli, something previous cognitive models have not addressed. In this paper, we examine whether recent machine learning approaches can help explain various behavioral phenomena from the psychological literature on cross‐situational word learning. We consider two variants of a multimodal neural network architecture and look at seven different phenomena associated with cross‐situational word learning and word learning more generally. Our results show that these networks can learn word‐referent mappings from a single epoch of training, mimicking the amount of training commonly found in cross‐situational word learning experiments. Additionally, these networks capture some, but not all of the phenomena we studied, with all of the failures related to reasoning via mutual exclusivity. These results provide insight into the kinds of phenomena that arise naturally from relatively generic neural network learning algorithms, and which word learning phenomena require additional inductive biases.
Purpose: To extract key learning from intervention studies in which qualitative aspects of dosage, dose form, have been examined for children with developmental language disorder (DLD) - in vocabulary, morphosyntax and phonology domains. This research paper emerged from a pair of systematic reviews, aiming to synthesise available evidence regarding qualitative and quantitative aspects of dosage respectively. Whilst quantitative aspects had been experimentally manipulated, the available evidence for dose form (tasks or activities within which teaching episodes are delivered) was less definitive. Despite this, the review uncovered insights of value to DLD research.
Method: A pre-registered systematic review (PROSPERO ID=CRD42017076663) adhering to PRISMA guidelines was completed. Included papers were: Quasi-experimental, RCT or cohort analytic studies, published in any language between January 2006 and May 2019; oral language interventions with vocabulary, morpho-syntax or phonology outcomes; and participants with DLD (M=3-18 years). The intention was to include papers in which dose form was experimentally manipulated or statistically analysed, while quantitative dosage aspects were controlled, such that definitive conclusions about optimal dose form could be drawn, and gaps in the evidence identified.
Results: 224 papers met the above inclusion criteria; 27 focused on dose form. No study controlled for all quantitative aspects of dosage such that we could effectively address our original research questions. Despite this, key points of learning emerged with implications for future research
Conclusions: There is tentative evidence of advantages for explicit over implicit instruction, and of the benefits of variability in input, elicited production and gestural and other visual supports. With careful design of dose form, there is potential to design more efficient interventions. SLP research would benefit from an agreed taxonomy of dose form components and standardised reporting of intervention studies to enable cross-study comparisons and a systematic accrual of knowledge to identify optimal dose form for clinical application.
There are two common challenges in particle swarm optimization (PSO) research, that is, selecting proper exemplars and designing an efficient learning model for a particle. In this article, we propose a triple archives PSO (TAPSO), in which particles in three archives are used to deal with the above two challenges. First, particles who have better fitness (i.e., elites) are recorded in one archive while other particles who offer faster progress, called profiteers in this article, are saved in another archive. Second, when breeding each dimension of a potential exemplar for a particle, we choose a pair of elite and profiteer from corresponding archives as two parents to generate the dimension value by ordinary genetic operators. Third, each particle carries out a specific learning model according to the fitness of its potential exemplars. Furthermore, there is no acceleration coefficient in TAPSO aiming to simplify the learning models. Finally, if an exemplar has excellent performance, it will be regarded as an outstanding exemplar and saved in the third archive, which can be reused by inferior particles aiming to enhance the exploitation and to save computing resources. The experimental results and comparisons between TAPSO and other eight PSOs on 30 benchmark functions and four real applications suggest that TAPSO attains very promising performance in different types of functions, contributing to both higher solution accuracy and faster convergence speed. Furthermore, the effectiveness and efficiency of these new proposed strategies are discussed based on extensive experiments.
Learning words from ambiguous naming events is difficult. In such situations, children struggle with not attending to task irrelevant information when learning object names. The current study reduces the problem space of learning names for object categories by holding color constant between the target and other extraneous objects. We examine how this influences two types of word learning (retention and generalization) in both 30-month-old children (Experiment 1) and the iCub humanoid robot (Experiment 2). Overall, all children and iCub performed well on the retention trials, but they were only able to generalize the novel names to new exemplars of the target categories if the objects were originally encountered in sets with objects of the same colors, not if the objects were originally encountered in sets with objects of different colors. These data demonstrate that less information presented during the learning phase narrows the problem space and leads to better word learning success for both children and iCub. Findings are discussed in terms of cognitive load and desirable difficulties.
How do children learn to link word to world? This question continues to be the subject of vibrant debate and research. Key insights into the processes of lexical development have steadily grown, beginning with the work of Roger Brown. This chapter highlights infants’ abilities to carve the sound stream into linguistic units and the event stream into non-linguistic units, so that word-referent mappings can be learned. High quality input, both linguistic (e.g., contingent conversations between parent and child) and non-linguistic (e.g., joint attention toward objects and events), is essential for the development of these foundational abilities and for word learning. We also discuss current theories tackling the so-called “mapping problem.” These proposals are beginning to converge on the idea that the processes children use to discover word meanings are complex, involving the use of associative, social, and linguistic cues, and change over time.
Semantically rich learning contexts facilitate semantic, phonological, and articulatory aspects of word learning in children with typical development (TD). However, because children with autism spectrum disorder (ASD) show differences at each of these processing levels, it is unclear whether they will benefit from semantic cues in the same manner as their typical peers. The goal of this study was to track how the inclusion of rich, sparse, or no semantic cues influences semantic, phonological, and articulatory aspects of word learning in children with ASD and TD over time. Twenty-four school-aged children (12 in each group), matched on expressive vocabulary, participated in an extended word learning paradigm. Performance on five measures of learning (referent identification, confrontation naming, defining, phonetic accuracy, and speech motor stability) were tracked across three sessions approximately one week apart to assess the influence of semantic richness on extended learning. Results indicate that children with ASD benefit from semantically rich learning contexts similarly to their peers with TD; however, one key difference between the two groups emerged - the children with ASD showed heightened shifts in speech motor stability. These findings offer insights into common learning mechanisms in children with ASD and TD, as well as pointing to a potentially distinct speech motor learning trajectory in children with ASD, providing a window into the emergence of stereotypic vocalizations in these children.
A nap soon after encoding leads to better learning in infancy. However, whether napping plays the same role in preschoolers' learning is unclear. In Experiment 1 (N = 39), 3-year-old habitual and nonhabitual nappers learned novel verbs before a nap or a period of wakefulness and received a generalization test examining word extension to novel actors after 24 hr. Only habitual and nonhabitual nappers who napped after learning generalized 24 hr later. In Experiment 2 (N = 40), children learned the same verbs but were tested within 2–3 min of training. Here, habitual and nonhabitual nappers retained the mappings but did not generalize. The results suggest that naps consolidate weak learning that habitual and nonhabitual nappers would otherwise forget over periods of wakefulness.
This study set out to examine whether shyness, an aversion to novelty and unfamiliar social situations, can affect the processes that underlie early word learning. Twenty-four-month-old children ( n =32) were presented with sets of one novel and two familiar objects, and it was found that shyer children were less likely to select a novel object as the referent of a novel label. Furthermore, not-shy children then showed evidence of retaining these novel mappings, but shy children did not. These findings suggest that shy children's aversion to novelty and to the unfamiliar context can impact on their word learning.
Understanding the factors which affect the age of acquisition (AoA) of words and concepts is fundamental to understanding cognitive development more broadly. Traditionally, studies of AoA have taken two approaches, either exploring the effect of linguistic variables such as input frequency (e.g., Naigles and Hoff-Ginsberg, 1998) or the semantics of the underlying concept, such as concreteness or imageability (e.g., Bird et al., 2001). Embodied theories of cognition, meanwhile, assume that concepts, even relatively abstract ones, can be grounded in the embodied experience. While the focus of such discussions has been mainly on grounding in external modalities, more recently some have argued for the importance of interoceptive features, or grounding in complex modalities such as social interaction. In this paper, we argue for the integration and extension of these two strands of research. We demonstrate that the psycholinguistic factors traditionally considered to determine AoA are far from sufficient to account for the variability observed in AoA data. Given this gap, we propose groundability as a new conceptual tool that can measure the degree to which concepts are grounded both in external and, critically, internal modalities. We then present a mechanistic theory of conceptual representation that can account for groundability in addition to the existing variables argued to influence concept acquisition in both the developmental and embodied cognition literatures, and discuss its implications for future work in concept and cognitive development.
A large literature has documented that comparison and contrast lead to better performance in a variety of tasks. However, studies of comparison and contrast present contradictory conclusions as to when and how these processes benefit learners. Across four studies, we examined how the specifics of the comparison and contrast task affect performance by systematically manipulating the feature variation and category structure in a category extension task performed by 3-year-old and 4-year-old children. Studies 1 (n = 48, M = 42.6 months) and 2 (n = 48, M = 42. 4 months) investigated comparison and contrast with high-density categories. Studies 3A (n = 60, M = 43.47 months), 3B (n = 48, M = 53. 2 months) and 4 (n = 48, M = 53.7 months) investigated comparison and contrast with low-density categories. Results indicated both category structure and feature variation affect the efficacy of comparison and contrast. Copyright © 2012 John Wiley & Sons, Ltd.
This paper investigates whether preverbal children form categories at different levels of abstraction in any specific sequence. In a longitudinal study, 20 infants were each tested twice, at 8 and 12 months of age. Half of the children solved a global-level task (animals-furniture), followed by a basic-level task (either dogs-birds, or chairs-tables) during each session. The other half received the basic-level task only. During familiarisation, all infants freely explored a series of four different exemplars from the same category presented one at a time. Infants saw all objects twice, for a total of eight trials. During the test phase, a new exemplar from the familiar category was presented, followed by a different-category exemplar. At 8 months of age, children discriminated between categories in the global-level task, but failed to do so in the basic-level task. At 12 months of age, infants recognised a category change in the basic-level task, but treated both test items as equally new in the global-level task. These findings support the hypothesis that infants younger than 1 year of age show a global-to-basic-level shift in category formation.
Young children learn words from a variety of situations, including shared storybook reading. A recent study by Horst et al. (2011a) demonstrates that children learned more new words during shared storybook reading if they were read the same stories repeatedly than if they were read different stories that had the same number of target words. The current paper reviews this study and further examines the effect of contextual repetition on children's word learning in both shared storybook reading and other situations, including fast mapping by mutual exclusivity. The studies reviewed here suggest that the same cognitive mechanisms support word learning in a variety of situations. Both practical considerations for experimental design and directions for future research are discussed.
Classic approaches to word learning emphasize referential ambiguity: In naming situations, a novel word could refer to many possible objects, properties, actions, and so forth. To solve this, researchers have posited constraints, and inference strategies, but assume that determining the referent of a novel word is isomorphic to learning. We present an alternative in which referent selection is an online process and independent of long-term learning. We illustrate this theoretical approach with a dynamic associative model in which referent selection emerges from real-time competition between referents and learning is associative (Hebbian). This model accounts for a range of findings including the differences in expressive and receptive vocabulary, cross-situational learning under high degrees of ambiguity, accelerating (vocabulary explosion) and decelerating (power law) learning, fast mapping by mutual exclusivity (and differences in bilinguals), improvements in familiar word recognition with development, and correlations between speed of processing and learning. Together it suggests that (a) association learning buttressed by dynamic competition can account for much of the literature; (b) familiar word recognition is subserved by the same processes that identify the referents of novel words (fast mapping); (c) online competition may allow the children to leverage information available in the task to augment performance despite slow learning; (d) in complex systems, associative learning is highly multifaceted; and (e) learning and referent selection, though logically distinct, can be subtly related. It suggests more sophisticated ways of describing the interaction between situation- and developmental-time processes and points to the need for considering such interactions as a primary determinant of development. (PsycINFO Database Record (c) 2012 APA, all rights reserved).
In 2 experiments, adults and children were tested in an object-selection task that examined whether Ss would (1) map a novel word onto a previously unnamed object and (2) extend the newly learned word to another exemplar. Exp 3 was a control study. Ss overwhelmingly selected the novel object as the referent for the novel term, even though the new label was never explicitly linked to the novel object. Ss also extended the new term and allowed it to preempt yet another novel label from applying to the just-named object. The existence of several lexical principles and the power of indirect word learning is supported. (PsycINFO Database Record (c) 2012 APA, all rights reserved)
The acquisition of word meaning is often partially attributed to fast mapping. However, recent research suggests that fast mapping and word learning may represent distinct components of language acquisition. Here we examine their interaction with a Hebbian Normalized Recurrence Network, a connectionist architecture that captures both online processing and long-term statistical learning. After training on a small lexicon, the model performed well above chance on a fast mapping task. Careful analyses of the weight changes, however, suggest that the fast mapping task can be solved with minimal learning. Thus, this model not only captures both long-term learning and online processes, but also provides unique insights regarding the relationship between fast mapping and word learning and that the two should be carefully distinguished.
The current study examines how focusing children's attention immediately after fast mapping improves their ability to retain novel names. Previous research suggests that young children can only retain novel names presented via referent selection if ostensive naming is provided and that such explicit naming works by increasing children's attention to the target and decreasing their attention to the competitor objects (Horst and Samuelson, 2008). This explanation of the function of ostensive naming after referent selection trials was tested by drawing 24-month-old children's attention to the target either by illuminating the target, covering the competitors, or both. A control group was given a social pragmatic cue (pointing). Children given social pragmatic cue support did not demonstrate retention. However, children demonstrated retention if the target object was illuminated, and also when it was illuminated and the competitors simultaneously dampened. This suggests that drawing children's attention to the target object in a manner that helps focus children's attention is critical for word learning via referent selection. Directing attention away from competitors while also directing attention toward a target also aids in the retention of novel words.
Upon fast mapping, children rarely retain new words even over intervals as short as 5 min. In this study, we asked whether the memory process of encoding or consolidation is the bottleneck to retention. Forty-nine children, mean age 33 months, were exposed to eight 2- or-3-syllable nonce neighbors of words in their existing lexicons. Didactic training consisted of six exposures to each word in the context of its referent, an unfamiliar toy. Productions were elicited four times: immediately following the examiner's model, and at 1-min-, 5-min-, and multiday retention intervals. At the final two intervals, the examiner said the first syllable and provided a beat gesture highlighting target word length in syllables as a cue following any erred production. The children were highly accurate at immediate posttest. Accuracy fell sharply over the 1-min retention interval and again after an additional 5 min. Performance then stabilized such that the 5-min and multiday posttests yielded comparable performance. Given this time course, we conclude that it was not the post-encoding process of consolidation but the process of encoding itself that presented the primary bottleneck to retention. Patterns of errors and responses to cueing upon error suggested that word forms were particularly vulnerable to partial decay during the time course of encoding.
Research suggests that variability of exemplars supports successful object categorization; however, the scope of variability's support at the level of higher-order generalization remains unexplored. Using a longitudinal study, we examined the role of exemplar variability in first- and second-order generalization in the context of nominal-category learning at an early age. Sixteen 18-month-old children were taught 12 categories. Half of the children were taught with sets of highly similar exemplars; the other half were taught with sets of dissimilar, variable exemplars. Participants' learning and generalization of trained labels and their development of more general word-learning biases were tested. All children were found to have learned labels for trained exemplars, but children trained with variable exemplars generalized to novel exemplars of these categories, developed a discriminating word-learning bias generalizing labels of novel solid objects by shape and labels of nonsolid objects by material, and accelerated in vocabulary acquisition. These findings demonstrate that object variability leads to better abstraction of individual and global category organization, which increases learning outside the laboratory.
To investigate whether phonological or semantic encoding cues improved the fast mapping or word learning performance of preschoolers with specific language impairment (SLI) or typical development (TD) and whether performance varied for words containing high- or low-frequency sublexical sequences that named familiar or unfamiliar objects.
Forty-two preschoolers with SLI, 42 preschoolers with TD matched for age and gender to the children with SLI, and 41 preschoolers with TD matched for expressive vocabulary and gender to the children with SLI learned words in a supported learning context. Fast mapping, word learning, and post-task performance were assessed.
Encoding cues had no effect on fast mapping performance for any group or on the number of words children learned to comprehend. Encoding cues appeared to be detrimental to word production for children with TD. Across groups, a clear learning advantage was observed for words with low-frequency sequences and, to a lesser extent, words associated with an unfamiliar object.
The results suggest that phonotactic probability and previous lexical knowledge affect word learning in similar ways for children with TD and SLI and that encoding cues were not beneficial for any group.
This study experimentally tested the relationship between children's lexicon size and their ability to learn new words within the domain of color. We manipulated the size of 25 20-month-olds' color lexicons by training them with two, four, or six different color words over the course of eight training sessions. We subsequently tested children's ability to extend new color words to new instances. We found that training with a broader number of color words led to increased extension of new words. The results suggest that children's learning history predicts their ability to learn new words within domains.
Two experiments explored the ability of 18-month-old infants to form an abstract categorical representation of tight-fit spatial relations in a visual habituation task. In Experiment 1, infants formed an abstract spatial category when hearing a familiar word (tight) during habituation but not when viewing the events in silence or when hearing a novel word. In Experiment 2, infants were given experience viewing and producing tight-fit relations while an experimenter labeled them with a novel word. Following this experience, infants formed the tight-fit spatial category in the visual habituation task, particularly when hearing the novel word again during habituation. Results suggest that even brief experience with a label and tight-fit relations can aid infants in forming an abstract categorical representation of tight-fit relations.
The spacing effect describes the robust phenomenon whereby memory is enhanced when learning events are distributed, instead of being presented in succession. We investigated the effect of spacing on children's memory and category induction. Three-year-old children were presented with two tasks, a memory task and a category induction task. In the memory task, identical instances of an object were presented and then tested in a multiple choice test. In the category induction task, different instances of a category were presented and tested in a multiple choice test. In both tasks, presenting the instances in a spaced sequence resulted in more learning than presenting the instances in a massed sequence, despite the difficulty created by the spaced sequence. The spaced sequence increased the difficulty of the task by allowing children time to forget the previous instance during the spaced interval.
Comparison mechanisms have been implicated in the development of abstract, relational thought, including object categorization. D. Gentner and L. L. Namy (1999) found that comparing 2 perceptually similar category members yielded taxonomic categorization, whereas viewing a single member of the target category elicited shallower perceptual responding. The present experiments tested 2 predictions that follow from Gentner and Namy's (1999) model: (a) Comparison facilitates categorization only when the targets to be compared share relational commonalities, and (b) providing common labels for targets invites comparison, whereas providing conflicting labels deters it. Four-year-olds participated in a forced-choice task. They viewed 2 perceptually similar target objects and were asked to "find another one." Results suggest an important role for comparison in lexical and conceptual development.
In two experiments we examined the relation between productive vocabulary and categorization at the basic and superordinate levels. Experiment 1 assessed categorization using a spontaneous object-manipulation task; Experiment 2 used a structured task employing 3 different levels of verbal support. In both experiments 24-month-old children participated in two sessions, with different levels of categorization tested at each session. Each child's productive vocabulary was assessed using either the Reznick & Goldsmith (1989) short parental report or the MacArthur Communicative Development Inventory. In both experiments, children performed equally well at the basic and superordinate levels, regardless of the task or level of verbal support. In Experiment 1 productive vocabulary was correlated with categorization at the basic level: children with larger productive vocabularies were more successful on basic-level tasks. The same relation between productive vocabulary and basic-level categorization was seen in the lowest verbal support condition of Experiment 2. There was no relation between vocabulary score and categorization at the superordinate level in either experiment. Possible implications of these findings are discussed.
Although vocabulary acquisition requires children learn names for multiple things, many investigations of word learning mechanisms teach children the name for only one of the objects presented. This is problematic because it is unclear whether children's performance reflects recall of the correct name-object association or simply selection of the only object that was singled out by being the only object named. Children introduced to one novel name may perform at ceiling as they are not required to discriminate on the basis of the name per se, and appear to rapidly learn words following minimal exposure to a single word. We introduced children to four novel objects. For half the children, only one of the objects was named and for the other children, all four objects were named. Only children introduced to one word reliably selected the target object at test. This demonstration highlights the over-simplicity of one-word learning paradigms and the need for a shift in word learning paradigms where more than one word is taught to ensure children disambiguate objects on the basis of their names rather than their degree of salience.
Does making an inference lead to better learning than being instructed directly? Two experiments evaluated preschoolers' ability to learn new words, comparing their memory for words learned via inference or instruction. On Inference trials, one familiar and one novel object was presented and children were asked to Point at the [object name (i.e., pizer)]. These trials required the child to infer that the novel label referred to the novel object and not to the familiar object. On Instruction trials, a novel object label directly referred to a novel object (e.g., This is a glark) and no familiar distracter object was shown. We found that although children looked longer at the novel target on Instruction trials, they showed poorer retention of the newly learned label compared to words learned on Inference trials. Hence, we found that inferential learning was superior to instruction. Relevance for optimal learning contexts and education are discussed.
Recent research demonstrates that both real-time variability in perceptual input and task demands influence young children’s word learning and categorisation. The current study extends these findings by testing both children and a dynamic field theory (DFT) computational model in a category labelling task. Specifically, children and the model were introduced to multiple category members that were either moderately or highly variable. Both children and the model were better able to learn category labels when the individual category members were moderately variable. Overall, these findings have implications for both our understanding of children’s categorization and the use of computational models to investigate cognition more generally.
Children learn words with remarkable speed and flexibility. However, the cognitive basis of young children’s word learning is disputed. Further, although research demonstrates that children’s categories and category labels are interdependent, how children learn category labels is also a matter of debate. Recently, biologically plausible, computational simulations of children’s behavior in experimental tasks have investigated the cognitive processes that underlie learning. The ecological validity of such models has been successfully tested by deploying them in robotic systems (Morse, Belpaeme, Cangelosi, & Smith, 2010). The authors present a simulation of children’s behavior in a word learning task (Twomey & Horst, 2011) via an embodied system (iCub; Metta, et al., 2010), which points to associative learning and dynamic systems accounts of children’s categorization. Finally, the authors discuss the benefits of integrating computational and robotic approaches with developmental science for a deeper understanding of cognition.
During the second year of life, infants develop a preference to attach novel labels to novel objects. This behavior is commonly known as “mutual exclusivity” (Markman, 198914.
Markman , E. M. 1989 . Categorization and naming in children: Problems of induction. , Cambridge, MA : MIT Press . View all references). In an intermodal preferential looking experiment with 19.5- and 22.5-month-olds, stimulus repetition was critical for observing mutual exclusivity. On the first occasion that a novel label was presented with 1 familiar object and 1 novel object, looking behavior was unsystematic. However, on reexposure to the same stimuli, 22.5-month-olds looked preferentially at the novel object prior to the re-presentation of the novel label. These findings suggest a powerful memory mechanism for novel labels and objects, enabling mutual exclusivity to emerge across repeated exposures to potential referents.
Conducted 9 experiments with a total of 663 undergraduates using the technique of priming to study the nature of the cognitive representation generated by superordinate semantic category names. In Exp I, norms for the internal structure of 10 categories were collected. In Exps II, III, and IV, internal structure was found to affect the perceptual encoding of physically identical pairs of stimuli, facilitating responses to physically identical good members and hindering responses to identical poor members of a category. Exps V and VI showed that the category name did not generate a physical code (e.g., lines or angles), but rather affected perception of the stimuli at the level of meaning. Exps VII and VIII showed that while the representation of the category name which affected perception contained a depth meaning common to words and pictures which enabled Ss to prepare for either stimulus form within 700 msec, selective reduction of the interval between prime and stimulus below 700 msec revealed differentiation of the coding of meaning in preparation for actual perception. Exp IX suggested that good examples of semantic categories are not physiologically determined, as the effects of the internal structure of semantic categories on priming (unlike the effects for color categories) could be eliminated by long practice. (57 ref) (PsycINFO Database Record (c) 2012 APA, all rights reserved)
A series of 3 experiments are reviewed in which infants between 4 and 10 months of age were familiarized with members of 2 basic-level object categories. The degree of distinctiveness between categories was varied. Preference tests were intended to determine whether infants formed a single category representation (at a more global level) or 2 basic-level representations. Across 3 experiments, 10-month-old infants appeared to have formed multiple basic-level categories, whereas younger infants tended to form broader, more inclusive representations. The tendency to form multiple categories was influenced to some extent by category distinctiveness. Whereas 10-month-olds formed separate categories for all contrasts, 7-month-olds did so only when the 2 familiarized categories were from separate global domains. A perceptual account of the global-to-basic shift in early categorization is offered. Task dependencies in early categorization are also discussed.
Four experiments explored the processes that bridge between referent selection and word learning. Twenty-four-month-old infants were presented with several novel names during a referent selection task that included both familiar and novel objects and tested for retention after a 5-min delay. The 5-min delay ensured that word learning was based on retrieval from long-term memory. Moreover, the relative familiarity of objects used during the retention test was explicitly controlled. Across experiments, infants were excellent at referent selection, but very poor at retention. Although the highly controlled retention test was clearly challenging, infants were able to demonstrate retention of the first 4 novel names presented in the session when referent selection was augmented with ostensive naming. These results suggest that fast mapping is robust for reference selection but might be more transient than previously reported for lexical retention. The relations between reference selection and retention are discussed in terms of competitive processes on 2 timescales: competition among objects on individual referent selection trials and competition among multiple novel name–object mappings made across an experimental session.
2 studies investigate whether 18-month-old children spontaneously sort objects into basic-level categories, and how this ability is related to naming. In Study 1, 18-month-old children were given spontaneous sorting tasks, involving both identical objects and objects with basic-level intracategory variation. Children were scored as having passed the tasks if they produced “exhaustive grouping,” that is, physically grouped all the objects of one kind into one location and the objects of the other kind into a different location. The children also received means-ends and object-permanence tasks. Children's parents received a checklist of early names. Children who produced exhaustive grouping used significantly more names than those who did not, in both identical and basic-level cases. There was no such relation between object-permanence and naming or between means-ends performance and naming. In Study 2, children received arrays of the same objects, with either identical objects or objects with basic-level variation in each group. No significant differences were found between the identical and basic-level tasks. However, as in the previous task, performance on both types of categorization was related to naming. Children who produced exhaustive grouping were reported to produce more names than those who did not. There appears to be a close relation between object categorization and naming in young children. The theoretical implications of this empirical association are discussed.
Toddlers' acquisition of the Novel Name–Nameless Category (N3C) principle was examined to investigate the developmental lexical principles framework and the applicability of the specificity hypothesis to relations involving lexical principles. In Study 1, we assessed the ability of 32 children between the ages of 16 and 20 months to use the N3C principle (operationally defined as the ability to fast map). As predicted, only some of the children could fast map. This finding provided evidence for a crucial tenet of the developmental lexical principles framewor: Some lexical principles are not available at the start of language acquisition. Children who had acquired the N3C principle also had significantly larger vocabularies and were significantly more likely to demonstrate 2-category exhaustive sorting abilities than children who had not acquired the principle. The 2 groups of children did not differ in either age or object permanence abilities. The 16 children who could not fast map were followed longitudinally until they attained a vocabulary spurt; at that time, their ability to fast map was retested (Study 2). Results provided a longitudinal replication of the findings of Study 1. Implications of these findings for both the developmental lexical principles framework and the specificity hypothesis are discussed.
Recent research demonstrated that although twenty-four month-old infants do well on the initial pairing of a novel word and novel object in fast-mapping tasks, they are unable to retain the mapping after a five-minute delay. The current study examines the role of familiarity with the objects and words on infants' ability to bridge between the initial fast mapping of a name and object, and later retention in the service of slow mapping. Twenty-four-month-old infants were familiarized with either novel objects or novel names prior to the referent selection portion of a fast-mapping task. When familiarized with the novel objects, infants retained the novel mapping after a delay, but not when familiarized with the novel words. This suggests familiarity with the object versus the word form leads to differential encoding of the name-object link. We discuss the implications of this finding for subsequent slow mapping.
Three experiments directly compared infants' categorization in variations of the visual familiarization task. In each experiment, 4- or 6-month-old infants were familiarized with a collection of dogs or cats and then their response to novel dogs and cats was assessed. In Experiment 1, 4-month-old infants responded to the exclusive distinction of dogs or cats when tested in a paired-comparison task. In Experiments 2 and 3, 6-month-old infants, but not 4-month-old infants, responded to this same distinction in a successive presentation task, even when the amount of familiarization was equated to that of the paired comparison task. Therefore, familiarization with a particular set of stimuli does not induce infants to respond to a single category but rather they respond to different categories depending on features of the task.
Two experiments were conducted to investigate the representation of categories of kitchen and bathroom things in 14- and 20-month-old children. These contextual categories are based on spatiotemporal relatedness rather than on the similarity of form or function that underlies the more frequently studied taxonomic categories. In the first experiment, a preferential-looking technique was used that demonstrated sensitivity to a kitchen category but not to a bathroom category. In the second experiment, an object-manipulation task was used that demonstrated sensitivity to both categories. The object-manipulation task was also used to show the presence of a basic-level category of cars and a more global category of animals. The data indicate that a variety of categories are represented at an early age.
Recent research on children's word learning has led to a paradox. Although word learning appears to be a deep source of insight into conceptual knowledge for children, preschoolers often categorize objects on the basis of shallow perceptual features such as shape. The current studies seek to resolve this discrepancy. We suggest that comparing multiple instances of a category enables children to extract deeper relational commonalities among category members. We examine 4-year-olds' categorization behaviors when asked to select a match for a target object (e.g., an apple) between a perceptually similar, out-of-kind object (e.g., a balloon) and a perceptually different category match (e.g., a banana). Children who learn a novel word as a label for multiple instances of the category are more likely to select the category match over the perceptual match. Children who learn a label for only one instance are equally likely to select either alternative. This effect is present even when individual target instances are more perceptually similar to the perceptual choice than to the category choice. We conclude that structural alignment processes may be important in the development of category understanding.
What mechanism implements the mutual exclusivity bias to map novel labels to objects without names? Prominent theoretical accounts of mutual exclusivity (e.g., Markman, 1989, 1990) propose that infants are guided by their knowledge of object names. However, the mutual exclusivity constraint could be implemented via monitoring of object novelty (see Merriman, Marazita, & Jarvis, 1995). We sought to discriminate between these contrasting explanations across two preferential looking experiments with 22-month-olds. In Experiment 1, infants viewed three objects: one name-known, two name-unknown. Of the two name-unknown objects, one was novel, and the other had been previously familiarized. The infants responded to hearing a novel label by increasing attention only to the novel, name-unknown object. In a second experiment in which the name-known object was absent, a novel label increased infants' attention to a novel object beyond baseline preference for novelty. The experiments provide clear evidence for a novelty-based mechanism. However, differences in the time course of disambiguation across experiments suggest that novelty processing may be influenced by contextual factors.
Determining the referent of a novel name is a critical task for young language learners. The majority of studies on children's referent selection focus on manipulating the sources of information (linguistic, contextual and pragmatic) that children can use to solve the referent mapping problem. Here, we take a step back and explore how children's endogenous biases towards novelty and their own familiarity with novel objects influence their performance in such a task. We familiarized 2-year-old children with previously novel objects. Then, on novel name referent selection trials children were asked to select the referent from three novel objects: two previously seen and one completely novel object. Children demonstrated a clear bias to select the most novel object. A second experiment controls for pragmatic responding and replicates this finding. We conclude, therefore, that children's referent selection is biased by previous exposure and children's endogenous bias to novelty.
Previous research suggests that competition among the objects present during referent selection influences young children's ability to learn words in fast mapping tasks. The present study systematically explored this issue with 30-month-old children. Children first received referent selection trials with a target object and either two, three or four competitor objects. Then, after a short delay, children were tested on their ability to retain the newly fast-mapped names. Overall, the number of competitors did not affect children's ability to form the initial name-object mappings. However, only children who encountered few competitors during referent selection demonstrated significant levels of retention. Results and implications are discussed in terms of the role of competition in studies of children's fast mapping. The relationship between referent selection and full word learning is also discussed.